Engineering insect-resistant plants by transgenic expression of an insecticidal spider-venom peptide

Abstract

The insecticidal spider-venom peptide o-hexatoxin-Hv1a (Hv1a) from the Australian Blue Mountains funnel-web spider Hadronyche versuta is one of the most potent insect-specific neurotoxins isolated to date. Hv1a blocks voltage-gated calcium channels in the insect central nervous system, a mechanism quite distinct from existing chemical insecticides. It induces a slow-onset paralysis that precedes death in a taxonomically wide range of insects. Hv1a's broad spectrum of target insects, novel mode of action, and absence of toxicity to vertebrates makes the Hv1a gene an attractive tool for generating insect resistant transgenic crops. The oral activity of Hv1a can be enhanced by coupling it to the plant lectin Galanthus nivalis agglutinin (GNA) or with the minor capsid protein of pea enation mosaic virus (CP). Recombinant fusions of Hv1a with GNA were produced using the Pichia pastoris expression system to study the intrinsic insecticidal activity of Hv1a-GNA and GNA-Hv1a fusion proteins. By using injection bioassays with houseflies, we found that the intrinsic insecticidal activity of Hv1a was maintained when it was fused to GNA. Moreover, feeding bioassays with diamondback moth larvae revealed that fusion of Hv1a to GNA, in either orientation, enhances its oral insecticidal activity. In order to generate transgenic plants expressing Hv1a alone or fused to GNA or CP, transformation vectors were constructed by ligating synthesised genes in the pAOV binary vector with the constitutive Cauliflower Mosaic Virus 35S promoter (35S) or the phloem tissue-specific Arabidopsis thaliana SUCROSE TRANSPORTER 2 (SUC2) promoter. Homozygous transgenic Arabidopsis were generated using the floral dip method of Agrobacterium-mediated plant transformation and subsequent herbicide selection. PCR of genomic DNA and western blotting were used to confirm integration of the transgenes and protein expression in the transgenic Arabidopsis respectively. Initial experiments revealed a very high level of mortality of Helicoverpa armigera larvae on wild-type plants due to the presence of endogenous glucosinolates, which masked the insecticidal effects of the transgenes Thus, a new set of transgenic plants was generated using an Arabidopsis cyp79B2 cyp79B myb28 myb29 quadruple mutant that lacks endogenous glucosinolates. Bioassays revealed that H. armigera larvae had a lower level of survival and retarded growth when fed on leaves of transgenic Arabidopsis expressing Hv1a toxins under 35S promoter control compared with those fed on gluc-null control plants. Moreover, larval mortality was higher for plants expressing Hv1a/GNA fusions than those expressing Hv1a or GNA alone. The highest larval mortality, lowest larval weight gain, and lowest level of leaf damage were observed for larvae fed on plants expressing GNA-Hv1a. Mortality was extremely high (~90%) for larvae fed on GNA-Hv1a plants for 15 days. The resistance to cotton bollworms conferred by expression of GNA-Hv1a in transgenic Arabidopsis highlights the potential of Hv1a transgenes as an alternative to harmful chemical insecticides. Moreover, Hv1a transgenes might provide a useful adjunct or alternative to Bt crops, and they might be useful for trait stacking with Bt transgenes.